Pulmonary embolism: non-diagnostic ventilation-perfusion scans: further investigation was required.

Clinical bottom line (level 4)

  1. About 4% of patients with suspected pulmonary embolism, good cardiorespiratory function and a non-diagnostic ventilation-perfusion scan developed a DVT or PE (NNF = 27 for 3 months) . This was not significantly different from patients with a confirmed PE who were anticoagulated.
  2. About 2% of patients with suspected pulmonary embolism, good cardiorespiratory function and a non-diagnostic ventilation-perfusion scan, who had serial negative impedance plethysmography and were not anticoagulated developed a DVT or PE. This was not significantly different from patients who did not have a PE.
Hull et al: Archives of Internal Medicine 1994; 154: 289-297
Expires September 2003

The study

Prospective cohort study with objective outcomes, not adjusted for confounding factors, not validated in an independent set of patients.

Setting: two tertiary hospitals, Canada

1564 patients (aged range 12 to 96 years; mean 56, 63%% female) suspected pulmonary embolism

Excluded if
  • impedance plethysmography or ventilation-perfusion scan could not be performed


    • All patients had ECG, chest x-ray, impedance plethysmography and a ventilation-perfusion scan. If IPG was abnormal they had a venogram. All investigations were reviewed by two independent observers.

    100% followed for 3 months
    Outcomes studied:
  • negative v/q scan: recurrent DVT or PE Recurrent PE: diagnosed by v/q scan: if not high-probability had pulmonary angiogram; Recurrent DVT diagnosed by IPG or venogram
  • negative v/q scan: fatal PE
  • non-diagnostic v/q scan: recurrent DVT or PE
  • non-diagnostic v/q scan: fatal PE
  • non-diagnostic v/q scan, leg testing negative after 14 days: recurrent DVT or PE
  • non-diagnostic v/q scan, leg testing negative after 14 days: fatal PE
  • high-probability v/q scan: recurrent DVT or PE
  • high-probability v/q scan: fatal PE

    • 67% of patients were out-patients.
    • Divided into three groups:
      • patients with high-probability v/q scan, who were anticoagulated for 3 months
      • patients with a normal v/q scan, who were not anticoagulated
      • patients with a non-diagnostic v/q scan, normal IPG and normal cardiorespiratory reserve. All had IPG on day 1, 3, 5, 7, 10 and 14. If IPG remained negative, anticoagulation was withheld. If it became positive, patients had a venogram and were anticoagulated if necessary
    • Patients with a non-diagnostic v/q scan, and decreased cardiorespiratory reserve had pulmonary angiography and were anticoagulated if PE was detected. Cardiorespiratory reserve was assessed at presentation or 10 days later. Decreased reserve if any of:
      • pulmonary oedema
      • right ventricular failure
      • systolic blood pressure <90
      • acute tachyarrhythmias
      • respiratory failure: FEV 1 <1.0; FVC <1.5; pO 2 <50 mmHg on air; pCO 2 >45 mmHg on air)

    The evidence

    outcome time to outcome number of patients/total number %
    (95% CI)
    negative v/q scan: recurrent DVT or PE 3 months / 0.7%
    (0.2% to 1.3%)
    negative v/q scan: fatal PE 3 months / 0.0%
    (0.0% to 0.51%)
    non-diagnostic v/q scan: recurrent DVT or PE 3 months / 4.4%
    (2.7% to 5.9%)
    non-diagnostic v/q scan: fatal PE 3 months / 0.0%
    (0.0% to 0.46%)
    non-diagnostic v/q scan, leg testing negative after 14 days: recurrent DVT or PE 3 months / 1.9%
    (0.8% to 3.0%)
    non-diagnostic v/q scan, leg testing negative after 14 days: fatal PE 3 months / 0.16%
    (0.05% to 0.47%)
    high-probability v/q scan: recurrent DVT or PE 3 months / 5.5%
    (1.8% to 9.2%)
    high-probability v/q scan: fatal PE 3 months / 0.69%
    (0.0% to 2.0%)

    • recurrent DVT or PE:
      • non-diagnostic v/q scan
      RR compared to negative scan group (95% CI)
      • 6.39 (2.25 to 18.1)
      NNF for three months (95% CI)
      • 27 (9 to 120)
      • non-diagnostic v/q scan, leg testing negative after 14 days
      RR
      • 2.80 (0.91 to 8.64)
      NNF
      • 81 (-1600 to 19)
      • high-probability v/q scan
      RR
      • 8.08 (2.47 to 26.5)
      NNF
      • 21 (6 to 100)
    • 46% of v/q scans non-diagnostic: similar to PIOPED study (41%).
    • Serial IPG scanning diagnosed 16 DVTs. (60% of total recurrent VTEs in non-diagnostic group).

    Comments

    1. Patients in non-diagnostic v/q, serial leg scanning group were older, more likely to be male, and had more CCF, COPD, pneumonia, or abnormal chest examination than other groups. All reasons for v/q scan to be non-diagnostic. Also increased chance of misdiagnosis of PE - this may explain why most patients with non-diagnostic scans have a similar outcome to patients without PE.
    2. The study was too small to give any indication about the effect on fatal PEs.
    3. The study suggests that patients with non-diagnostic scans need further investigation to rule out small chance of venous thromboembolism (84/711 i.e 12% in this study).

    Citation

    1. Hull RD, Raskob GE, Ginsberg JS, et al: A noninvasive strategy for the treatment of patients with suspected pulmonary embolism. Archives of Internal Medicine 1994; 154: 289-297
    Contributor: Chris Ball and Clare Wotton, September 2000
    Reviewer:

    Clinical Question.
    Patient suspected PE
    Intervention or Exposure non-diagnostic ventilation-perfusion scan
    Outcome DVT or PE